community pool / model answer

Muse-Glimmer-30B-GGUF

chat · unsloth/Muse-Glimmer-30B-GGUF

33.3 tok/s · model median · 12 runs in the model record.

Reference rig	Quantization	Median result	Basis	TTFT (ms)	Peak VRAM	Max context	Engines
Arc Pro B70 32GB ×4	8-bit	161.9 tok/s	reported	2,872	- GB	32,768	llama.cpp
Radeon AI Pro R9700 32GB	4-bit	33.2 tok/s	reported	478	- GB	2,048	llama.cpp
Radeon AI Pro R9700 32GB	5-bit	25.8 tok/s	reported	559	- GB	2,048	llama.cpp
RTX 3060 12GB ×2	4-bit	37.8 tok/s	reported	1,388	17.3 GB	65,536	llama.cpp
RTX 3080 12GB	3-bit	13.8 tok/s	reported	2,811	9.4 GB	16,384	llama.cpp
RTX 3090 24GB ×2	6-bit	31.7 tok/s	reported	176	- GB	2,048	llama.cpp
RTX 4060 Ti 16GB ×2	6-bit	44 tok/s	reported	-	- GB	125,000	llama.cpp
RTX PRO 6000 Blackwell 96GB	8-bit	48.1 tok/s	reported	402	32.8 GB	131,072	llama.cpp

Context tested

The table above is the tested context for each exact rig and quantization cell. Empty fields remain empty rather than being filled with an estimate.